AI-assisted review of implanted devices before a revision procedure is seeing promise.
By now, everyone with access to the news has seen some impact of artificial intelligence (AI) and machine learning (ML) on the world. Everything from legal briefs to pop hits is being created using AI tools, which leaves many wondering about the near-term impact on healthcare (Table 1). Work is being done to address the use of these technologies, such as President Biden’s “Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence”1 and FDA’s white paper titled, “Artificial Intelligence & Medical Products: How CBER, CDER, CDRH, and OCP are Working Together.”2 Documents like these are helping developers take some simple first steps on a long road.
FDA has seen a huge increase in the rate of AI/ML submissions, with a more than 39% rise in submissions from 2019 to 2020.3 While data indicates the submission rate is slowing, FDA has predicted the new submissions will still reach a growth rate over 30%.3
Some of these tools have already started to be a part of the administrative side of healthcare4 but it is still unknown how quickly AI and ML tools will be a regular part of the day-to-day clinical environment, supporting diagnosis and treatment. Work completed by MedTech Strategies indicates clinicians are excited about the real-world use of AI in their clinics, suites, and operating rooms but are still cautious about workflow and reimbursement coverage. Concerns regarding AI causing more work—time needed to double-check the outputs or add “clicks” to their software applications, for example—will be with us as these new technologies are adopted.
Of the recent set of new AI/ML devices cleared (i.e., Jan. 1 through July 31, 2023), there was a large representation in radiology with 79% of submissions in this area,3 which makes sense given the clear connection of imaging data to software applications (Table 2), but what does this mean for orthopedics? As of Oct. 19, 2023, FDA published an update to the AI/ML-Enabled Medical Devices list.3 This list included 692 devices submitted, however, only one of the devices listed was reviewed by an orthopedic panel—NuVasive Inc.’s Pulse System. Many companies are looking at AI applications for a wide range of uses in orthopedic applications.
AI-assisted review of implanted devices before a revision procedure is seeing promise. Data published in the journal Arthroplasty indicated that a deep learning model was able to predict the type of unique knee arthroplasty implants with 99% accuracy, which could result in significant time savings for pre-op planning and efficiency during the procedure.5
By Ilsa Webeck | Orthopedics Design & Technology
Image Credit: Ilsa Webeck | ODT
Be the first to comment